Neurophysiological Effects of Whole Coffee Cherry Extract in Older Adults with Subjective Cognitive Impairment: A Randomized, Double-Blind, Placebo-Controlled, Cross-Over Pilot Study

Abstract

Bioactive plant-based compounds have shown promise as protective agents across multiple domains including improvements in neurological and psychological measures. Methodological challenges have limited our understanding of the neurophysiological changes associated with polyphenol-rich supplements such as whole coffee cherry extract (WCCE). In the current study, we (1) compared 100 mg of WCCE to a placebo using an acute, randomized, double-blind, within-subject, cross-over design, and we (2) conducted a phytochemical analysis of WCCE. The primary objective of the study was to determine the neurophysiological and behavioral changes that resulted from the acute administration of WCCE. We hypothesized that WCCE would increase brain-derived neurotrophic factor (BDNF) and glutamate levels while also increasing neurofunctional measures in cognitive brain regions. Furthermore, we expected there to be increased behavioral performance associated with WCCE, as measured by reaction time and accuracy. Participants underwent four neuroimaging scans (pre- and post-WCCE and placebo) to assess neurofunctional/metabolic outcomes using functional magnetic resonance imaging and magnetic resonance spectroscopy. The results suggest that polyphenol-rich WCCE is associated with decreased reaction time and may protect against cognitive errors on tasks of working memory and response inhibition. Behavioral findings were concomitant with neurofunctional changes in structures involved in decision-making and attention. Specifically, we found increased functional connectivity between the anterior cingulate and regions involved in sensory and decision-making networks. Additionally, we observed increased BDNF and an increased glutamate/gamma-aminobutyric acid (GABA) ratio following WCCE administration. These results suggest that WCCE is associated with acute neurophysiological changes supportive of faster reaction times and increased, sustained attention.

Keywords: 7T; functional magnetic resonance imaging; nutraceuticals; polyphenols; spectroscopy.

Conflict of interest statement

This work was supported by VDF FutureCeuticals, Inc. (grant #G00010324, awarded to J.L.R. and D.T.B.). Study capsules were provided by VDF FutureCeuticals, Inc. Z.J.P., B.V.N., and J.M.H. are employees and receive compensation from VDF FutureCeuticals, Inc. Following completion of the study, data analysis, and manuscript write-up, J.L.R. has since engaged in a paid consultant role with VDF FutureCeuticals, Inc., however all analyses and manuscript development/editing occurred prior to the establishment of this relationship. All other authors report no conflicts of interest. VDF FutureCeuticals, Inc. maintained the blind drug assignment, but otherwise played no part in the data collection or analysis. The sponsor provided minor proofreading comments on the final draft of the manuscript prior to submission, with ultimate approval of any changes to the manuscript contents by all authors unaffiliated with the sponsor.

Figures

Figure 1

LC–MS base peak chromatogram (BPC) profile of whole coffee cherry extract (WCCE), adapted from the work of Nemzer and colleagues (under review) [27]. The major detected peaks were labelled with peak numbers ranging from 1 to 24, and the compounds corresponding to each peak are identified here: 1. gluconic acid; 2. quinic acid; 3. malic acid; 4. citric acid; 5. 2-hydroxyglutaric acid; 6. 3-O-caffeoylquinic acid (3-CQA); 7. protocatechualdehyde; 8. 3- coumaroylquinic acid (3-CoQA); 9. 5-CQA; 10. 3-feruloylquinic acid (3-FQA); 11. 4-CQA; 12. caffeic acid; 13. 5-CoQA; 14. 4-CoQA; 15. 5-FQA; 16. 4-FQA; 17. quinic acid-glucoside-R*; 18. 3-dicaffeoylquinic acid (3-DiCQA); 19. 5-DiCQA; 20. 4-DiCQA; 21. 3-Caffeoyl-5-FQA; 22. valeroylquinic acid (VQA) diglucoside-R*; 23. caffeoyl tryptophan; and 24. dimethylcaffeic acid. Other compounds identified in WCCE in the positive ion mode include pantothenic acid, trigonelline, choline, and glycerophosphocholine derivatives. Figure adapted from the work of Nemzer et al. (under review) [27]. A listing of the typical polyphenols found in WCCE can be found in Table 1.

Figure 2

Study design overview. For the Visit 1 and Visit 2 Assessment, 0:00 indicates the participants arrival at the imaging center. Subsequent times indicate hours and minutes since arrival. AUMRIRC: Auburn University MRI Research Center.

Figure 3

Magnetic resonance spectroscopy (MRS) voxel placement in the dorsal anterior cingulate cortex (dACC) and representative spectra. The 7 Tesla (7T) spectroscopy offers significant advantages, including more accurate assessments of glutamate and glutamine.

Figure 4

Differences in the pre–post consumption of WCCE or placebo for go trials (uppermost panel), no-go trials (middle panel), and all trials collapsed (bottom panel). In all panels, WCCE is presented on the left side, and the placebo is presented on the right. Statistic images were thresholded on magnitude (z ≥ 2.3), as well as cluster extent-determined by z > 2.3 and a corrected cluster significance threshold of p < 0.05. Local maxima tables for each contrast are available at https://osf.io/qypr8/. Abbreviations: LACC = left anterior cingulate cortex; LIFG = left inferior frontal gyrus; LINS = left insula; LMOG = left middle occipital gyrus; LPCC = left posterior cingulate cortex; RCG = right cingulate gyrus; and STG = superior temporal gyrus.

Figure 5

Anterior cingulate functional connectivity differences (post > pre) during the n-back task. WCCE showed greater connectivity post-consumption between the ACC and the paracingulate, the precuneus, and portions of the superior frontal gyrus and frontal poles. The placebo demonstrated a greater connectivity between the ACC and portions of the left dorsolateral prefrontal cortex, as well as the precuneus. Data were thresholded with puncorrected < 0.01 height threshold, pFDR-corrected < 0.05 cluster threshold, one-tailed.

Figure 6

Bar graph with M ± SEM demonstrating the changes in glutamate, gamma-aminobutyric acid (GABA), and the ratios of glutamate/GABA (GLU/GABA) and glutamine/GABA (GLN/GABA) within the anterior cingulate cortex following the administration of either the placebo or WCCE. Y-axis units of measurement are institutional units. A * indicates significance at the p < 0.05 level for one-tailed t-tests. For descriptive data, please see the supplemental table hosted at https://osf.io/qypr8/. Clear bars represent the placebo, while gray bars represent WCCE. Striped bars indicate post-measurements.

Figure 7

Pre- and post-consumption data for exosomal brain-derived neurotrophic factor (BDNF). Thin black lines indicate individual participant data, while the thicker white line depicts the group average. The dark gray shading surrounding the white line uses locally weighted regression (loess) for a visualization of group variability. Line graphs were created with open-source software ggplot2 in R.